Inhibition of the hERG potassium channel by phenanthrene: a polycyclic aromatic hydrocarbon pollutant

Figures

• 

  Fig. 1. A Sequence alignments of zERG and hERG showing very high homology in the pore helix, S5 and S6 regions that contribute to the canonical drug binding site and residues investigated here for phenanthrene. The structural elements of the channel pore domain are coloured to match the structural figure in Fig. 6a. B Upper traces show Bi Control IhERG, Bii IhERG during application of 10 µM phenanthrene (Phen), Biii Current during washout of Phen. IhERG was elicited by a voltage protocol shown as lower traces, comprised of a 2 s depolarizing pulse to +20 mV, followed by repolarization to −40 mV. C Time course of IhERG inhibition: plots of IhERG tails in control, during phenanthrene application and washout. D Concentration response relation for phenanthrene inhibition of IhERG. Six different concentrations between 1 µM and 100 µM were tested (1 µM, 5 µM, 10 µM, 30 µM, 50 µM and 100 µM). At least 5 replicates per concentration were obtained. The data were fitted with Eq. 2. Phenanthrene inhibited IhERG carried by the hERG1a isoform with a half-maximal inhibitory concentration (IC50) of 17.6 ± 1.7 µM and Hill coefficient of 0.94 ± 0.09

• 

  Fig. 2. A Bar charts comparing deactivation parameters for IhERG (hERG1a) tails elicited on repolarization to −40 mV from +20 mV in Control (black) and 10 µM phenanthrene (grey). Biexponential fitting yielded fast and slow deactivation time constants (τf and τs respectively), plotted in Ai and Aii. Aiii shows that the proportion of fast deactivating current increased in phenanthrene. Significance is denoted by asterisks **p < 0.01; *p < 0.05 (n = 5). Bar charts show mean ± SEM values, with individual experimental values in control and phenanthrene superimposed as filled circles and squares respectively. B: Bi shows representative traces of hERG1a/1b IhERG elicited by the standard protocol shown in Fig. 1 in the absence and presence of 10 µM phenanthrene. Bii shows concentration response data for 5 phenanthrene concentrations (1 µM, 5 µM, 10 µM, 30 µM; 100 µM; n = at least 5 at each concentration). The data were fitted with Eq. 2, yielding a half-maximal inhibitory concentration (IC50) 1.8 ± 0.3 µM and Hill coefficient of 0.80 ± 0.09. C Bar charts showing deactivation parameters for IhERG (hERG1a/1b) tails elicited on repolarization to -40 mV from +20 mV in Control (black) and 1 µM…

• 

  Fig. 3. A Representative IhERG traces (upper) are shown for Ai: in control solution and Aii: in 10 µM phenanthrene. IhERG was elicited by a voltage protocol (lower traces) that was comprised of 2 s command pulses applied at 10 mV increments between −40 and +60 mV. Only selected traces are shown for clarity of display. Each current trace is labelled with its corresponding test potential. B Mean I–V relation for end-pulse currents. The currents were normalized to the maximum current in each of control and 10 µM phenanthrene. Corresponding mean fractional inhibition by 10 µM Phenanthrene is also shown. C Mean IhERG tails in control (black) and phenanthrene (grey) following the different voltage commands. Currents were normalized to the maximum current in each of control and 10 µM phenanthrene Boltzmann fitting gave an activation V0.5 for Control of 1.3 ± 1.6 mV, slope: 7.2 ± 0.2 (n = 16) and for phenanthrene gave a V0.5 of −8.9 ± 1.5 (p < 0.001 vs Control), Slope: 7.8 ± 0.3 (n = 16)

• 

  Fig. 4. A Upper traces show representative IhERG records elicited by protocol shown as lower traces. Tail currents were elicited at −40 mV following depolarizing pulses to +40 mV for differing time periods between 50 and 1100 ms. Left hand panel shows recordings in control solution and right-hand panel contains recordings from the same cell during exposure to 30 µM phenanthrene. B Plot of mean fractional block of IhERG tail by 30 µM phenanthrene against command pulse duration (n = 5). A monoexponential fit to these data (from each experiment) gave an inhibition time constant (τinhib) of 130.2 ± 15.4 ms. A one-way ANOVA showed an overall significance in the time dependence of inhibition (p < 0.0001), with (Bonferroni) post hoc testing showing that inhibition following pulses of up to 200 ms in duration was significantly different from all durations of 500 ms and longer (p < 0.05 or less for all comparisons)

• 

  Fig. 5. Effect of hERG mutations on the inhibitory effect of phenanthrene on IhERG. A–G show representative tail current traces (upper) for control IhERG (black) and during application of 30 µM Phen IhERG (grey), elicited by same voltage protocol used for Fig. 1 (lower) as described. Ai: wild type (WT), Aii: N588K, B S624A, C Y652A, D F656V, E F656T, F F557L and G M651A. H Bar chart shows extent of mean tail current block by 30 µM for WT, N588K, Y652A, S624A, F656V, F656T, F557L and M651A IhERG respectively. One way ANOVA (F = 61.2, p < 0.0001). Significance for post hoc Bonferroni comparisons of each mutant with WT is denoted by asterisks ****p < 0.0001, ***p < 0.001 (WT n = 5, N588K n = 7, Y652A n = 6, S624A n = 5, F656V n = 5, F656T n = 5, F557L n = 7 and M651A n = 6). Bar charts show mean ± SEM values, with individual experimental values for each mutant superimposed as the different symbols overlying the plotted bars

• 

  Fig. 6. A Low energy score pose for phenanthrene (yellow space filling representation) docked into the hERG pore. The hERG subunit with docked phenanthrene shows the location within a single subunit of the mutations described in the text; the structural elements are coloured to match the alignment in Fig. 1a. Purple spheres are K+ ions in the 1 and 3 positions of the selectivity filter. B Close-up of a hydrophobic pocket containing docked phenanthrene (yellow sticks) in the same pose as panel A. An aromatic pi-stacking interaction with the F557 side chain and close interactions with the M651 side chain are indicated with dotted lines